
. 
. *       ********************************************************************* *
. *       Replication Code for Online Appendix for "`Because He's Gay': How Race, Gender, a
> nd Sexuality   
. *       Shape Perceptions of Judicial Fairness"
. *       Journal of Politics
. *       Authors: Ana Bracic, Mackenzie Israel-Trummel, Tyler Johnson, and Kathleen Tipler
. *       ********************************************************************* *
.  
.  
. *       ********************************************************************* *
. *       Analyses of Open Ended Statements Presented in Sections 4.4 and 4.5   *
. *       ********************************************************************* *
. 
. 
. *               ********************************************************************* *
. *               Analyses of statements are in the order they appear in the main text  *
. *               ********************************************************************* *
. 
. 
. *       ***************************************** *
. *               SECTION 4.4: Are the Choices Deliberate?  *      
. *       ***************************************** *
. 
. use conjoint_judges_open_ended.dta

. 
. 
. *To address the possibility of respondents complaining about the conjoint being a false c
> hoice, we looked at responses coded 88 on variable riddle and then we counted, by hand, t
> he number of responses that 1) commented on having to choose a judge and 2) commented tha
> t it was a coin toss. 
. 
. 
. #delimit;
delimiter now ;
. foreach var of varlist race gender sexuality party age lawschool previousjob generalminor
> ity refertolegalcase racist sexist homophobic ageist antiparty antischool antijob ideolog
> yopenliberal ideologyclosedmorals judgesameasme moreexperience impartial knowsdiscriminat
> ion mydesiredoutcome doesntfit riddle {;
  2.         tab `var';
  3.         };

       race |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,986       92.03       92.03
          1 |        345        7.97      100.00
------------+-----------------------------------
      Total |      4,331      100.00

     gender |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,851       88.92       88.92
          1 |        480       11.08      100.00
------------+-----------------------------------
      Total |      4,331      100.00

  sexuality |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,815       88.09       88.09
          1 |        516       11.91      100.00
------------+-----------------------------------
      Total |      4,331      100.00

      party |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,951       91.23       91.23
          1 |        380        8.77      100.00
------------+-----------------------------------
      Total |      4,331      100.00

        age |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,751       86.61       86.61
          1 |        580       13.39      100.00
------------+-----------------------------------
      Total |      4,331      100.00

 law school |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,965       91.55       91.55
          1 |        366        8.45      100.00
------------+-----------------------------------
      Total |      4,331      100.00

   previous |
        job |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,394       78.37       78.37
          1 |        937       21.63      100.00
------------+-----------------------------------
      Total |      4,331      100.00

    general |
   minority |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,280       98.82       98.82
          1 |         51        1.18      100.00
------------+-----------------------------------
      Total |      4,331      100.00

   refer to |
 legal case |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,044       93.37       93.37
          1 |        287        6.63      100.00
------------+-----------------------------------
      Total |      4,331      100.00

     racist |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,295       99.17       99.17
          1 |         36        0.83      100.00
------------+-----------------------------------
      Total |      4,331      100.00

     sexist |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,291       99.08       99.08
          1 |         40        0.92      100.00
------------+-----------------------------------
      Total |      4,331      100.00

 homophobic |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,255       98.27       98.27
          1 |         75        1.73      100.00
------------+-----------------------------------
      Total |      4,330      100.00

     ageist |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,307       99.45       99.45
          1 |         24        0.55      100.00
------------+-----------------------------------
      Total |      4,331      100.00

 anti-party |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,262       98.41       98.41
          1 |         69        1.59      100.00
------------+-----------------------------------
      Total |      4,331      100.00

anti-school |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,324       99.84       99.84
          1 |          7        0.16      100.00
------------+-----------------------------------
      Total |      4,331      100.00

   anti-job |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,242       97.95       97.95
          1 |         89        2.05      100.00
------------+-----------------------------------
      Total |      4,331      100.00

ideology/op |
 en/liberal |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,977       91.83       91.83
          1 |        354        8.17      100.00
------------+-----------------------------------
      Total |      4,331      100.00

ideology/cl |
osed/morals |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,263       98.43       98.43
          1 |         68        1.57      100.00
------------+-----------------------------------
      Total |      4,331      100.00

 judge same |
      as me |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,297       99.21       99.21
          1 |         34        0.79      100.00
------------+-----------------------------------
      Total |      4,331      100.00

       more |
 experience |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,994       92.22       92.22
          1 |        337        7.78      100.00
------------+-----------------------------------
      Total |      4,331      100.00

  impartial |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,863       89.19       89.19
          1 |        468       10.81      100.00
------------+-----------------------------------
      Total |      4,331      100.00

      knows |
discriminat |
        ion |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,085       94.32       94.32
          1 |        246        5.68      100.00
------------+-----------------------------------
      Total |      4,331      100.00

 my desired |
    outcome |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,232       97.71       97.71
          1 |         99        2.29      100.00
------------+-----------------------------------
      Total |      4,331      100.00

doesn't fit |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,829       88.43       88.43
          1 |        501       11.57      100.00
------------+-----------------------------------
      Total |      4,330      100.00

     riddle |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,004       92.45       92.45
         77 |        102        2.36       94.80
         88 |        176        4.06       98.87
         99 |         49        1.13      100.00
------------+-----------------------------------
      Total |      4,331      100.00

. #delimit cr     
delimiter now cr
. 
. 
. *From these tabulations we see that the most common attributes mentioned were prior occup
> ation (21.6%), age (13.4%), sexuality (11.9%), and gender (11.1%). About 7.6% of response
> s were nonsensical and 11.6% did not fit any of our coding categories. We also see that 5
> .7% of respondents chose a judge because they would be familiar with discrimination. Comb
> ining counts for ideologyopenliberal and ideologyclosedmorals we see that about 10% of re
> spondents invoked political motivation in their choice.
. 
.  
. *generating a count to see how many multiple categories there are and how many overall bl
> ank responses we have
. 
. gen total_reasons =.
(5,168 missing values generated)

. replace total_reasons = race + gender + sexuality + party + age + lawschool + previousjob
>  + generalminority + refertolegalcase + racist + sexist + homophobic + ageist + antiparty
>  + antischool + antijob + ideologyopenliberal +  ideologyclosedmorals+  judgesameasme+  m
> oreexperience + impartial + knowsdiscrimination +  mydesiredoutcome + doesntfit
(4,329 real changes made)

. 
. tab total_reasons

total_reaso |
         ns |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |        325        7.51        7.51
          1 |      2,478       57.24       64.75
          2 |        951       21.97       86.72
          3 |        376        8.69       95.40
          4 |        138        3.19       98.59
          5 |         45        1.04       99.63
          6 |         11        0.25       99.88
          7 |          4        0.09       99.98
          8 |          1        0.02      100.00
------------+-----------------------------------
      Total |      4,329      100.00

.  
. *Of the 4,331 decisions, 7.5% provided no justification. 
.  
.  
. *generating a count of statements that mention an identity
.   
. gen total_identity_reasons =.
(5,168 missing values generated)

. replace total_identity_reasons = race + gender + sexuality + generalminority 
(4,331 real changes made)

. 
. tab total_identity_reasons

total_ident |
ity_reasons |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      3,176       73.33       73.33
          1 |        949       21.91       95.24
          2 |        175        4.04       99.28
          3 |         31        0.72      100.00
------------+-----------------------------------
      Total |      4,331      100.00

. 
. *In 26.7% of answers respondents mentioned a judge’s race, gender, sexuality, or general 
> minority status
. 
. 
. *generating a count of intolerant statements
. 
. gen total_animus_reasons =.
(5,168 missing values generated)

. replace total_animus_reasons = racist + sexist + homophobic + ageist
(4,330 real changes made)

. 
. tab total_animus_reasons

total_animu |
  s_reasons |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |      4,161       96.10       96.10
          1 |        163        3.76       99.86
          2 |          6        0.14      100.00
------------+-----------------------------------
      Total |      4,330      100.00

. 
. *3.9% of statements were explicitly intolerant
. 
. 
. 
. *       ***************************************** *
. *               SECTION 4.5: Does Case Type Matter ?      *
. *       ***************************************** *
. 
. 
. *are people more likely to bring up race in the race case and so on?
. 
. *generating dummies for each case type
. 
. gen race_case =.
(5,168 missing values generated)

. replace race_case =1 if discrimtype =="racial"
(948 real changes made)

. replace race_case =0 if discrimtype !="racial"
(4,220 real changes made)

. replace race_case =. if pipe1 == ""
(420 real changes made, 420 to missing)

. 
. 
. gen gender_case =.
(5,168 missing values generated)

. replace gender_case =1 if discrimtype =="gender"
(950 real changes made)

. replace gender_case =0 if discrimtype !="gender"
(4,218 real changes made)

. replace gender_case =. if pipe1 == ""
(420 real changes made, 420 to missing)

. 
. gen gender_id_case =.
(5,168 missing values generated)

. replace gender_id_case =1 if discrimtype =="gender identity"
(950 real changes made)

. replace gender_id_case =0 if discrimtype !="gender identity"
(4,218 real changes made)

. replace gender_id_case =. if pipe1 == ""
(420 real changes made, 420 to missing)

. 
. gen sexuality_case =.
(5,168 missing values generated)

. replace sexuality_case =1 if discrimtype =="sexual orientation"
(949 real changes made)

. replace sexuality_case =0 if discrimtype !="sexual orientation"
(4,219 real changes made)

. replace sexuality_case =. if pipe1 == ""
(420 real changes made, 420 to missing)

. 
. 
. gen religion_case =.
(5,168 missing values generated)

. replace religion_case =1 if discrimtype =="religious"
(951 real changes made)

. replace religion_case =0 if discrimtype !="religious"
(4,217 real changes made)

. replace religion_case =. if pipe1 == ""
(420 real changes made, 420 to missing)

. 
. 
. *doing an independent groups t-test comparing each case to just the religion case; 
. *creating a variable where religion is default (0) and other case is 1;
. 
. gen religion_race_cases =.
(5,168 missing values generated)

. replace religion_race_cases = 0 if religion_case ==1
(951 real changes made)

. replace religion_race_cases = 1 if race_case ==1
(948 real changes made)

. label var religion_race_cases "0 if religion case, 1 if race case"

. 
. 
. gen religion_gender_cases =.
(5,168 missing values generated)

. replace religion_gender_cases = 0 if religion_case ==1
(951 real changes made)

. replace religion_gender_cases = 1 if gender_case ==1
(950 real changes made)

. label var religion_gender_cases "0 if religion case, 1 if gender case"

. 
. 
. gen religion_gender_id_cases =.
(5,168 missing values generated)

. replace religion_gender_id_cases = 0 if religion_case ==1
(951 real changes made)

. replace religion_gender_id_cases = 1 if gender_id_case ==1
(950 real changes made)

. label var religion_gender_id_cases "0 if religion case, 1 if gender identity case"

. 
. 
. gen religion_sexuality_cases =.
(5,168 missing values generated)

. replace religion_sexuality_cases = 0 if religion_case ==1
(951 real changes made)

. replace religion_sexuality_cases = 1 if sexuality_case ==1
(949 real changes made)

. label var religion_sexuality_cases "0 if religion case, 1 if sexuality case"

. 
. 
. 
. ttest gender, by(religion_gender_cases)

Two-sample t test with equal variances
------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
       0 |     868    .0679724    .0085481    .2518436    .0511949    .0847498
       1 |     867    .1649366    .0126113    .3713375    .1401843    .1896888
---------+--------------------------------------------------------------------
combined |   1,735    .1164265    .0077023    .3208282    .1013197    .1315334
---------+--------------------------------------------------------------------
    diff |           -.0969642    .0152321               -.1268394    -.067089
------------------------------------------------------------------------------
    diff = mean(0) - mean(1)                                      t =  -6.3658
Ho: diff = 0                                     degrees of freedom =     1733

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.0000         Pr(|T| > |t|) = 0.0000          Pr(T > t) = 1.0000

. *Footnote 11: 16.5% mention gender in the sex discrimination case compared to 6.8% in the
>  religion case (p < 0.05).
. 
. ttest gender, by(religion_gender_id_cases)

Two-sample t test with equal variances
------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
       0 |     868    .0679724    .0085481    .2518436    .0511949    .0847498
       1 |     863    .0996524    .0102022    .2997096    .0796282    .1196765
---------+--------------------------------------------------------------------
combined |   1,731    .0837666    .0066606    .2771175    .0707029    .0968303
---------+--------------------------------------------------------------------
    diff |             -.03168    .0133034               -.0577724   -.0055876
------------------------------------------------------------------------------
    diff = mean(0) - mean(1)                                      t =  -2.3814
Ho: diff = 0                                     degrees of freedom =     1729

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.0087         Pr(|T| > |t|) = 0.0174          Pr(T > t) = 0.9913

. *Footnote 12: 10% mention gender in the gender identity case compared to 6.8% in the reli
> gion case (p < 0.05). 
. 
. ttest sexuality, by(religion_gender_id_cases)

Two-sample t test with equal variances
------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
       0 |     868    .0702765    .0086811    .2557598    .0532382    .0873148
       1 |     863    .1853998    .0132365    .3888469    .1594202    .2113793
---------+--------------------------------------------------------------------
combined |   1,731    .1276719    .0080235    .3338205    .1119351    .1434087
---------+--------------------------------------------------------------------
    diff |           -.1151233    .0158112               -.1461343   -.0841123
------------------------------------------------------------------------------
    diff = mean(0) - mean(1)                                      t =  -7.2811
Ho: diff = 0                                     degrees of freedom =     1729

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.0000         Pr(|T| > |t|) = 0.0000          Pr(T > t) = 1.0000

. *Footnote 13: 18.5% mention sexuality in the gender identity case compared to 7% in the r
> eligion case (p < 0.05).  
. 
. ttest race, by(religion_race_cases)

Two-sample t test with equal variances
------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
       0 |     868    .0587558    .0079867    .2353026    .0430803    .0744313
       1 |     862    .1809745    .0131207    .3852205    .1552222    .2067267
---------+--------------------------------------------------------------------
combined |   1,730    .1196532    .0078053    .3246494    .1043443    .1349621
---------+--------------------------------------------------------------------
    diff |           -.1222187     .015336               -.1522978   -.0921397
------------------------------------------------------------------------------
    diff = mean(0) - mean(1)                                      t =  -7.9694
Ho: diff = 0                                     degrees of freedom =     1728

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.0000         Pr(|T| > |t|) = 0.0000          Pr(T > t) = 1.0000

. *Footnote 14: 18.1% mention race in the racial discrimination case compared to 5.8% in th
> e religion case (p < 0.05).
. 
. ttest sexuality, by(religion_sexuality_cases)

Two-sample t test with equal variances
------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
       0 |     868    .0702765    .0086811    .2557598    .0532382    .0873148
       1 |     871    .1745121    .0128679    .3797673    .1492563    .1997679
---------+--------------------------------------------------------------------
combined |   1,739    .1224842     .007864    .3279385    .1070603    .1379081
---------+--------------------------------------------------------------------
    diff |           -.1042356    .0155325               -.1346998   -.0737713
------------------------------------------------------------------------------
    diff = mean(0) - mean(1)                                      t =  -6.7108
Ho: diff = 0                                     degrees of freedom =     1737

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.0000         Pr(|T| > |t|) = 0.0000          Pr(T > t) = 1.0000

. *Footnote 15: 17.5% mention sexuality in the sexual orientation case compared to 7% in th
> e religion case (p < 0.05).
. 
. 
. ttest previousjob, by(religion_case)

Two-sample t test with equal variances
------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
       0 |   3,463    .2027144    .0068326    .4020795    .1893181    .2161107
       1 |     868    .2707373    .0150906    .4445968    .2411189    .3003557
---------+--------------------------------------------------------------------
combined |   4,331    .2163473    .0062574    .4118013    .2040796     .228615
---------+--------------------------------------------------------------------
    diff |           -.0680229    .0155989               -.0986048   -.0374411
------------------------------------------------------------------------------
    diff = mean(0) - mean(1)                                      t =  -4.3608
Ho: diff = 0                                     degrees of freedom =     4329

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(T < t) = 0.0000         Pr(|T| > |t|) = 0.0000          Pr(T > t) = 1.0000

. 
. *how many times does prior occupation appear in the religion case compared to other cases
> ? Footnote 16: 27.1% mention previous occupation in the religion case compared to 20.3% i
> n all other cases (p < 0.05).

